A recent epidemiological study and a 45 year follow up study of college students have emphasized the relation of insomnia and the use of sedative/hypnotics to mental health. This proposal is designed to further characterize the actions of benzodiazepine sedative/hypnotics, the most widely prescribed psychoactive drugs in the U. S. It is based on several puzzling findings in the sleep studies literature which have been widely reported but never explained, notably: 1. Benzodiazepine hypnotics decrease slow-wave sleep (SWS), which has traditionally been considered to be the "deepest" stage, yet these drugs leave the patient with the sense of having slept deeply and well. 2. At least half of normal volunteers and chronic insomniacs who are aroused by auditory stimuli during Stage 4 of SWS report that they believe they had previously been awake. This perception of being awake when aroused from Stage 4 occurs more frequently than from Stage 2 or REM. The present proposal is derived from preliminary data indicating that the benzodiazepine hypnotic triazolam 0.25 mg induces reports of having been asleep in Stage 4 sleep, and induces a description of "participating' rather than "watching" in mental activity during Stage 4. It is hypothesized that a drug effect on the perception of being awake or asleep during the night may influence the morning retrospective description of depth of sleep and it is suggested that a change in the perception of being awake or asleep may result from either drug-induced alterations in the form of mental activity or in the amount of SWS. These hypotheses will be tested by determining whether tolerance develops to this effect on perception of being awake or asleep, and whether it is dose-related. It will be determined whether a drug-induced alteration in perception of being awake or asleep occurs with two drugs which, in contrast to benzodiazepines, increase SWS, and which do (zopiclone) and do not (ritanserin) have clinical hypnotic properties. This proposal will clarify the poorly understood relationship between hypnotic effects on polygraphic measures and subjective reports of sleep, and also suggests that slow-wave sleep is much richer in cognitive activity than has previously been supposed.